In addition to being important for normal fetal and postnatal growth, there is increasing evidence that insulin-like growth factors I and II (IGF-I, IGF-II) also support the growth of certain cancers. Biologic responses to IGF-I and IGF-II are signaled by the IGF-I receptor. Therefore, we are focusing our research effort on understanding signaling by the IGF-I receptor. We have used the yeast two-hybrid system to identify new binding partners for the IGF-I receptor. Two binding partners which we identified in a screen of a human fetal brain library were two isoforms of 14-3-3 protein and SOCS-2 (supressor of cytokine signaling). To explore a functional role for 14-3-3 in IGF-I receptor signaling we overexpressed IGF-I receptors that were mutated in the binding sites for 14-3-3 as defined in the yeast two-hybrid system. Deletion of the COOH tail of the receptor beyond residue 1310 resulted in a marked increase in the number of large colonies in a soft agar assay. When this mutation was combined with a S1283A mutation the number of large colonies decreased significantly. The 14-3-3 binding mutant transfectants have been tested for tumorigenesis in immunodeficient mice. In agreement with the soft agar assay results, the 1310 deletion mutant transfectants produced tumors which appeared earlier and grew more rapidly than the 1310 deletion/S1283A transfectants. These results suggest that 14-3-3 may be involved in a transformation pathway emanating from the COOH tail of the receptor. To explore a functional role for SOCS-2 in IGF-I receptor signaling, we developed a tetracycline inducible system for SOCS-2 expression in MG-63 human osteosarcoma cells. We found that when SOCS-2 expression was induced, IGF-I stimulated DNA synthesis was inhibited by 50%, suggesting that SOCS-2 may also be a negative regulator of IGF-I receptor signaling. Induction of SOCS-2 resulted in a 50% inhibition of MAP kinase activation by IGF-I without a change in MAP kinase expression. We have also performed a yeast two-hybrid screen of a MG-63 human osteosarcoma cell library using the IGF-I receptor cytoplasmic domain as bait. In addition to eight proteins which were previously identified as binding partners for the IGF-I receptor, seven new interactors were found. Baserga and his colleagues reported that an embryo fibroblast cell line derived from an IGF-I receptor knockout mouse could not be transformed by a variety of oncoproteins including SV40 large T antigen as assessed by anchorage independent growth. To confirm this potentially important finding we developed mouse embryo fibroblast lines from wild type (3) and receptor knockout (3) embryos and attempted to transform these cell lines with SV40 large T antigen. In agreement with Baserga's findings, the three receptor negative cell lines initially could not be transformed by SV40 large T antigen as measured by formation of large colonies in soft agar. However, with continued passage of clones from one of the receptor negative transfectants, all three clones formed large colonies in soft agar. These SV40 T antigen transfectants also produced tumors in immunodeficient mice. Thus the requirement for the IGF-I receptor to be present in order to transform cells with SV40 large T antigen is not absolute.